Artificial neural network technique for rainfall forecasting applied to Alexandria, Egypt

Amr H. El-Shafie, A. El-Shafie, Hasan G. El Mazoghi, A. Shehata, Mohd. Raihan Taha

Research output: Contribution to journalArticle

29 Citations (Scopus)

Abstract

Two rainfall prediction models were developed and implemented in Alexandria, Egypt. These models are Artificial Neural Network ANN model and Multi Regression MLR model. A Feed Forward Neural Network FFNN model was developed and implemented to predict the rainfall on yearly and monthly basis. In order to evaluate the incomes of both models, statistical parameters were used to make the comparison between the two models. These parameters include the Root Mean Square Error RMSE, Mean Absolute Error MAE, Coefficient Of Correlation CC and BIAS. The data set that has been used in this study includes daily measurements for the rainfall and temperature and cover the period from 1957 to 2009. The FFNN model has shown better performance than the MLR model. The MLR model revealed a humble prediction performance. The linear nature of MLR model estimators makes it inadequate to provide good prognostics for a variable characterized by a highly nonlinear physics. On the other hand, the ANN model is a nonlinear mapping tool, which potentially is more suitable for rain (nonlinear physics) forecasts. More detailed studies are necessary due to uncertainties inherent in weather forecasting and efforts should be addressed to the problem of quantifying them in the ANN models.

Original languageEnglish
Pages (from-to)1306-1316
Number of pages11
JournalInternational Journal of Physical Sciences
Volume6
Issue number6
Publication statusPublished - Mar 2011
Externally publishedYes

Fingerprint

Egypt
forecasting
Rain
Neural networks
Physics
Weather forecasting
weather forecasting
income
performance prediction
Feedforward neural networks
physics
root-mean-square errors
rain
Mean square error
estimators
regression analysis

Keywords

  • Alexandria city
  • Multi regression
  • Neual network
  • Rainfall forecasting

ASJC Scopus subject areas

  • Physics and Astronomy(all)
  • Electronic, Optical and Magnetic Materials

Cite this

Artificial neural network technique for rainfall forecasting applied to Alexandria, Egypt. / El-Shafie, Amr H.; El-Shafie, A.; El Mazoghi, Hasan G.; Shehata, A.; Taha, Mohd. Raihan.

In: International Journal of Physical Sciences, Vol. 6, No. 6, 03.2011, p. 1306-1316.

Research output: Contribution to journalArticle

El-Shafie, AH, El-Shafie, A, El Mazoghi, HG, Shehata, A & Taha, MR 2011, 'Artificial neural network technique for rainfall forecasting applied to Alexandria, Egypt', International Journal of Physical Sciences, vol. 6, no. 6, pp. 1306-1316.
El-Shafie, Amr H. ; El-Shafie, A. ; El Mazoghi, Hasan G. ; Shehata, A. ; Taha, Mohd. Raihan. / Artificial neural network technique for rainfall forecasting applied to Alexandria, Egypt. In: International Journal of Physical Sciences. 2011 ; Vol. 6, No. 6. pp. 1306-1316.
@article{6ab7fc4ffc9c4bfa86b0457409d473cb,
title = "Artificial neural network technique for rainfall forecasting applied to Alexandria, Egypt",
abstract = "Two rainfall prediction models were developed and implemented in Alexandria, Egypt. These models are Artificial Neural Network ANN model and Multi Regression MLR model. A Feed Forward Neural Network FFNN model was developed and implemented to predict the rainfall on yearly and monthly basis. In order to evaluate the incomes of both models, statistical parameters were used to make the comparison between the two models. These parameters include the Root Mean Square Error RMSE, Mean Absolute Error MAE, Coefficient Of Correlation CC and BIAS. The data set that has been used in this study includes daily measurements for the rainfall and temperature and cover the period from 1957 to 2009. The FFNN model has shown better performance than the MLR model. The MLR model revealed a humble prediction performance. The linear nature of MLR model estimators makes it inadequate to provide good prognostics for a variable characterized by a highly nonlinear physics. On the other hand, the ANN model is a nonlinear mapping tool, which potentially is more suitable for rain (nonlinear physics) forecasts. More detailed studies are necessary due to uncertainties inherent in weather forecasting and efforts should be addressed to the problem of quantifying them in the ANN models.",
keywords = "Alexandria city, Multi regression, Neual network, Rainfall forecasting",
author = "El-Shafie, {Amr H.} and A. El-Shafie and {El Mazoghi}, {Hasan G.} and A. Shehata and Taha, {Mohd. Raihan}",
year = "2011",
month = "3",
language = "English",
volume = "6",
pages = "1306--1316",
journal = "International Journal of Physical Sciences",
issn = "1992-1950",
publisher = "Academic Journals",
number = "6",

}

TY - JOUR

T1 - Artificial neural network technique for rainfall forecasting applied to Alexandria, Egypt

AU - El-Shafie, Amr H.

AU - El-Shafie, A.

AU - El Mazoghi, Hasan G.

AU - Shehata, A.

AU - Taha, Mohd. Raihan

PY - 2011/3

Y1 - 2011/3

N2 - Two rainfall prediction models were developed and implemented in Alexandria, Egypt. These models are Artificial Neural Network ANN model and Multi Regression MLR model. A Feed Forward Neural Network FFNN model was developed and implemented to predict the rainfall on yearly and monthly basis. In order to evaluate the incomes of both models, statistical parameters were used to make the comparison between the two models. These parameters include the Root Mean Square Error RMSE, Mean Absolute Error MAE, Coefficient Of Correlation CC and BIAS. The data set that has been used in this study includes daily measurements for the rainfall and temperature and cover the period from 1957 to 2009. The FFNN model has shown better performance than the MLR model. The MLR model revealed a humble prediction performance. The linear nature of MLR model estimators makes it inadequate to provide good prognostics for a variable characterized by a highly nonlinear physics. On the other hand, the ANN model is a nonlinear mapping tool, which potentially is more suitable for rain (nonlinear physics) forecasts. More detailed studies are necessary due to uncertainties inherent in weather forecasting and efforts should be addressed to the problem of quantifying them in the ANN models.

AB - Two rainfall prediction models were developed and implemented in Alexandria, Egypt. These models are Artificial Neural Network ANN model and Multi Regression MLR model. A Feed Forward Neural Network FFNN model was developed and implemented to predict the rainfall on yearly and monthly basis. In order to evaluate the incomes of both models, statistical parameters were used to make the comparison between the two models. These parameters include the Root Mean Square Error RMSE, Mean Absolute Error MAE, Coefficient Of Correlation CC and BIAS. The data set that has been used in this study includes daily measurements for the rainfall and temperature and cover the period from 1957 to 2009. The FFNN model has shown better performance than the MLR model. The MLR model revealed a humble prediction performance. The linear nature of MLR model estimators makes it inadequate to provide good prognostics for a variable characterized by a highly nonlinear physics. On the other hand, the ANN model is a nonlinear mapping tool, which potentially is more suitable for rain (nonlinear physics) forecasts. More detailed studies are necessary due to uncertainties inherent in weather forecasting and efforts should be addressed to the problem of quantifying them in the ANN models.

KW - Alexandria city

KW - Multi regression

KW - Neual network

KW - Rainfall forecasting

UR - http://www.scopus.com/inward/record.url?scp=79957945389&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79957945389&partnerID=8YFLogxK

M3 - Article

VL - 6

SP - 1306

EP - 1316

JO - International Journal of Physical Sciences

JF - International Journal of Physical Sciences

SN - 1992-1950

IS - 6

ER -